Respiratory failure remains a major cause of mortality in the trauma patient. The combination of an episode of shock and tissue damage followed by sepsis, frequently results in a severe form of acute lung injury; the injury from both sepsis and trauma appearing to be greater in both frequency and in severity from that due to sepsis alone. The major characteristics of this lung injury in man and animals remains that of pulmonary artery hypertension and increased protein permeability leading to pulmonary edema. We have recently demonstrated that increased transvascular fluid flux seen after hemorrhagic shock alone is basically that produced by the shock and resuscitation induced hypoproteinemia. We now want to define the mechanism of injury of the other components, namely sepsis and the effect of soft tissue damage on the sepsis injury. Recent findings indicate that three factors are involved: (1) the cyclooxygenase products of arachidonic acid with the role of the lipoxygenase products still undetermined, (2) products released from sequestered neutrophils, specifically oxygen radicals, and (3) platelets possibly via serotonin release. We plan to implement physiological, immunological and biochemical methods already being used by the four established investigators to define this relationship. The animal model will continue to be the unanesthetized sheep with lung and soft tissue lymph fistulae using lymph flow and protein content to reflect QF and protein permeability and the content of mediators in lymph to reflect that released by or into the lung or from a focus of tissue damage. The septic episode will initially be endotoxin while a bacterial sepsis model is being developed. Tissue trauma will be produced by a full thickness burn (anesthetic to the animal). Neutrophil chemotactic factors, complement being one, and factors activating neutrophils to release O2 radicals, will be characterized as will the role of the pulmonary macrophage in initiating the process. Cyclooxygenase products will continue to be monitored as well as lipoxygenase products (initially by selective inhibition until an assay is available). The relationship of serotonin to the above will be determined. The role of oxygen radicals will be studied by monitoring the degree of lipid peroxidation present in lung tissue and its correlation with physiologic injury. Treatment modalities can be pursued once the mechanism of injury is better defined.